[GRLUG] (Linux || !Linux) WAS: NOT LINUX - IT

[Michael Mol]

On Mon, May 10, 2010 at 10:50 AM, Bob Kline <bob.kline at gmail.com> wrote:
> http://www.bloomberg.com/apps/harvardbusiness?sid=He84aa970dc941badd41e92799657d701
> There are a number of IT people on
> this list.  Does anything in the article
> above seem familiar?

Yeah, it reads like a Dilbert strip written by someone who doesn't
understand Dilbert.

> An aside, the current issue of Linux
> Journal agrees that "Linux" is the
> kernel.  In which case about 99.9%
> of what is discussed here is "NOT
> LINUX."
> i.e., it might be easier to write "LINUX"
> in the subject line on those rare
> occasions when that is in fact the topic.
>    -- Bob

By that logic, "PC" and "Computer" may as well be mutually synonymous
with Windows, in the mind of the vast majority of Americans. (I'm not
talking about limiting the demographic to 35-and-below!)

By that logic, "Coke" may as well be synonymous with "cola".
Actually, in some parts of the South, "Coke" is synonymous with
{soda|pop}.

By that logic, "Internet" was synonymous with "Internet Explorer" for
several years.

Linux is *trademarked*. What if someone took the Linux kernel (and
*only* the Linux kernel) and wrote a completely proprietary,
closed-source utilization infrastructure around it? Do you call the
whole resulting thing "Linux"? Do you just call the kernel Linux?

What if someone took the Linux kernel (and *only* the Linux kernel),
forked it, applied significant modifications, wrote their own software
stack to surround it, and open-sourced the whole thing? DO you call
the whole resulting thing Linux? Do you just call the kernel Linux? Is
even the *kernel* Linux, considering it's a fork?

We've sat in a software ecosystem for *years* where the infrastructure
around the kernel was by-and-large written by GNU, to the point where
we stopped bothering to call it "GNU/Linux", except in formal
situations, and even then only by people who particularly cared. Then
we started looking at situations (such as Linuxkernel+busybox, and
other micro environments) where it was still the core Linux kernel,
but with a different software stack.  Now we're looking at a situation
where the kernel was forked, and the software stack is distinct from
GNU, busybox or other stacks.

Generalization is OK, but there are contexts where distinction is important!

(This is going to be my only post on the subject today; late start at
work. Stuff to get done. IHBT'd, but at least I know when to quit...)

-- 
:wq